not greatly exceeding the cost of the present volume. Dr. Hyslop's critical introduction is a careful and suggestive piece of criticism; but for the very reason that it is a criticism rather than an exposition, it can hardly be of great service as an introduction to the text.
This book contains the first complete statement that Weismann has given us of his views on heredity and the hereditary substance. The "germ plasm" is here analyzed into its ultimate constituents, the "biophors." Groups of these minute vital units form "determinants," units of the second order, each of which is to control an independently variable cell or group of cells in the mature body. The determinants are themselves grouped into "ids," each id containing all the determinants necessary to produce a complete organism. Ontogeny takes place through the gradual breaking up of the ids in cell-division into simpler combinations of determinants, until finally every independently variable cell of the body is controlled by a single determinant, which stamps its character upon the cell by distributing its constituent biophors through the cell-substance.
To explain the phenomena of regeneration we are required to assume that the cells at each ontogenetic stage contain as inactive, "accessory idioplasm," the determinants of all succeeding stages. Moreover, part of the germ-plasm contained in the nucleus of the ovum remains unchanged during the whole of ontogeny, being passed through a definite series of cells, the "germ-track," to the point where it forms the nucleus of a germ-cell of the next generation.
The treatment of amphimixis contains nothing essentially new except the section on "The Struggle of the Ids in Ontogeny," where the share which each parent has in the determination of the offspring is held to depend on the success of the ids derived from that parent in obtaining control of the cells. The relative strength of the ids is measured partly by their rate of assimilation and multiplication, and partly by the number of "homologous" determinants they contain that are also "homodynamous," i.e., not only controlling the same cell, but impressing the same character upon it.
Two hypotheses suffice to explain the phenomena of reversion: first, that in the course of phylogeny all homologous determinants are not modified at once, so that the nucleus of a germ-cell may contain determinants in various stages of transformation; and second, that the "reducing division" may occur in such a way as occasionally to leave the older determinants in the majority. The theory reaches a rather appalling complexity, when, to account for alteration of generations and sexual dimorphism, we are obliged to assume that nearly all determinants in a germ-cell are double or multiple.
